1958 Journal o f Food Protection, Vol. 76, No. 11, 2013, Pages 1958-1962 doi: 10.4315/0362-028X.JFP-13-168 Copyright © , International Association for Food Protection
Research Note
Prevalence and Quantitative Detection of Salm onella in Retail Raw Chicken in Shaanxi, China JIAQI WANG,1 HAIYUN WU, 12 MIAO SONG,1 FENGQIN LI,3 JIANGHUI ZHU,3 MEILI XI,1 XIN WANG, 1 XIAODONG XIA, 1 JIANGHONG MENG, 14 BAOWEI YANG,1* AND SHENGHUI C M 5 1College o f Food Science and Engineering, Northwest A&F University, Shaanxi, 712100, China; 2School o f Pharmacy, Shanghai Jiao Tong University, Shanghai, 200240, China; 3China National Center fo r Food Safety Risk Assessment, Beijing, 100021, China; 4Joint Institute for Food Safety and Applied Nutrition, and Department o f Nutrition and Food Science, University o f Maryland, College Park, Maryland 20742, USA; and ^National Institutes fo r Food and Drug Control, Beijing, 100050, China MS 13-168: Received 26 April 2013/Accepted 28 June 2013
A BSTR A CT Quantitative Salmonella concentrations and prevalence of Salmonella in raw chicken at the retail level in Shaanxi province, China, was determined in this study. Two hundred forty samples were collected in Yangling and the surrounding cities, in Shaanxi Province, China, for data collection over 12 successive months from April 2011 to March 2012. During the whole surveillance year, the overall Salmonella contamination rate of retail raw chicken was identified as 43.3%, the mostprobable-number (MPN) values ranged from 0.0036 to 0.8596 MPN per g, and the average value was 0.1655 MPN per g, except in 11 of the samples, which had MPN values of more than 11 MPN per g. In April 2011, the highest prevalence rate, which was 80.0%, was observed, and the corresponding MPN value was 0.8596 MPN per g. Observed in November 2011, the lowest prevalence rate and the corresponding MPN value were 5.0% and 0.0036 MPN per g, respectively. Prevalence of Salmonella in chicken collected from the supermarket (49.2%) was higher but not significantly different from that in the wet markets (37.5%), although the MPN value of samples in the wet market (0.1912 MPN per g) was higher than that in the supermarket (0.1396 MPN per g). Prevalence of Salmonella was the highest in the frozen chicken (53.3%) compared with chilled chicken (45.0%) and freshly slaughtered chicken (37.5%); however, the MPN value of the freshly slaughtered chicken (0.1912 MPN per g) was higher than those of either frozen chicken (0.1804 MPN per g) or the chilled chicken (0.0988 MPN per g).
Salmonella is a genus of gram-negative rod-shaped bacteria of the Enterobacteriaceae family and can cause a wide range of human diseases such as gastroenteritis and enteric fever. The natural habitat of Salmonella is in the intestinal tract of animals and humans (19). Poultry and poultry products are usually incriminated in outbreaks of human salmonellosis (9). Chicken is widely acknowledged as a significant reservoir for Salmonella and has frequently been regarded as a source of Salmonella contamination (2). Salmonella often reaches the carcasses from the intestinal tracts or fecal materials of poultry (9). A link between the prevalence of Salmonella infection in poultry and human cases of salmonellosis was identified in a risk assessment conducted by the Food and Agriculture Organization of the United Nations and the World Health Organization (15). This report indicated that reducing rates of Salmonella infection in poultry flocks would lead to a direct reduction of salmonellosis within that community (15). Thus, surveillance of the level of Salmonella contam ination in food and food processing environments is * Author for correspondence. Tel: +86-29-87092486; Fax: 87092486; E-mail: [email protected].
+ 86-29-
necessary to control the spread of this pathogen from food to human. Control of Salmonella in food animals has been successful in the United States (8), Sweden, and Denmark (14) and has led to low levels of salmonellosis in these countries. The true incidence of salmonellosis in both humans and animals is difficult to evaluate in developing countries because of the lack of epidemiological surveil lance systems (12). In China, most research on Salmonella focused on the prevalence in different food animals, although few studies determined the quantitative Salmonella concentrations in food samples. The aim of the study was to determine the prevalence and quantitative concentrations of Salmonella in raw poultry at the retail level in Shaanxi province, China, and to collect data for quantitative risk assessment of Salmonella that was carried out by the China National Center for Food Safety Risk Assessment. M A TER IA LS A N D M E TH O D S Sample collection. For a total of 240 samples, 20 chicken carcasses, including 10 freshly slaughtered chickens in a wet market and 5 chilled and 5 frozen chicken samples in a supermarket, respectively, were collected monthly in Yangling and the surrounding cities in Shaanxi Province, China. After collection, each sample was placed aseptically in a sterile plastic
J. Food Prat., Vol. 76, No. 11
PREVALENCE AND MPN OF SAIMONF.UA IN RETAIL CHICKEN
1959
TABLE 1. Prevalence o f Salmonella in retail chicken by month Sampling date
No. of samples examined
April 2011 May 2011 June 2011 July 2011 August 2011 September 2011 October 2011 November 2011 December 2011 January 2012 February 2012 March 2012
20 20 20 20 20 20 20 20 20 20 20 20
Total
240
No. (%) of Salmonellapositive samples0 16 14 10 13 12 13 3 1 4 2 2 14
(80.0) B (70.0) b e (50.0) ABDE (65.0) BCE (60.0) ABCE (65.0) b c e (15.0) ACD (5.0) d (20.0) ADE (10.0) AD (10.0) AD (70.0) BE
104 (43.3)
Avg MPN value (MPN/g)
No. (%) of MPN value >11 (MPN/g)
0.8596 0.2453 0.0329 0.2084 0.3948 0.2104 0.0128 0.0036 0.0171 0.0046 0.0062 0.0805
2 (1 0 ) 4 (20) 0 (0 ) 3 (15) 0 (0) 1 (5) 0 (0 ) 0 (0 ) 0 (0 ) 0 (0 ) 0 (0 ) 1 (5)
0.1655
11 (4.58)
a Within a column, data labeled with different letters indicate significantly different (P < 0.05). bag. After it was marked, the sample was immediately stored in an icebox and transported to the Microbial Food Safety Laboratory at Northwest A&F University (Yangling, Shaanxi, China). The time from collection to isolation was no more than 4 h to avoid the potential growth of Salmonella during sample storage and transportation. The chicken type was acquired from the package labels or from the retailer.
Microbiological analysis. Analytical testing for Salmonella was carried out in the Microbial Food Safety Laboratory in Northwest A&F University. The analytical procedure for Salmo nella detection and most-probable-number (MPN) enumeration followed the method suggested by Food Safety and Inspection Service (FSIS) of the U.S. Department of Agriculture and China National Center for Food Safety Risk Assessment method (13). Briefly, each sample was weighed after transportation to the laboratory, then 500-fold of sterilized buffered peptone water (BPW; Difco, BD, Sparks, MD), according to the weight (kilograms), was added to the chicken carcass. The chicken was rinsed by manually massaging and using a rocking motion for 5 min to ensure that the surface, internal, and external layers of the chicken carcass would contact the rinse, and the rinse solution was used for the Salmonella concentration test. Three empty tubes were set up, as well as three tubes containing 9 ml of BPW and three tubes containing 9.9 ml of BPW. Ten milliliters of the chicken rinse solution was added to three empty tubes (representing 1 g), and then 1 ml of the rinse was transferred to each of three tubes that contained 9 ml of BPW (representing 0.1 g) and 100 pi of the rinse to each of three tubes that contained 9.9 ml of BPW (representing 0.01 g), respectively. If needed, 10-fold serial dilutions representing 0 .0 0 1-g samples were made. After inoculation, the tubes were preenriched for 20 to 24 h at 35 + 2°C. A portion (0.5 + 0.05 ml) of the preenrichment culture was transferred to 10 ml of tetrathionate broth (Difco) and 0.1 ± 0.02 ml into 10 ml of modified Rappaport-Vassiliadis broth (Difco), then incubated at 42 + 0.5°C with shaking at 100 rpm for 22 to 24 h. After incubation, a loop (diameter: 3 mm) of tetrathionate broth and modified Rappaport-Vassiliadis broth culture was streaked onto xylose lysine Tergitol 4 agar (Difco) and incubated at 35 + 2°C for 22 to 24 h. Colonies of presumptive Salmonella on xylose lysine Tergitol 4 plates were selected and purified on MacConkey agar; isolates with typical Salmonella phenotypes were confirmed by using the API 20E test kit (bioMerieux, Inc., Hazelwood, MO).
The MPN of each sample was acquired from the table of the MPN index, and 95% confidence limits for various combinations of positive tubes in a three-tube dilution series, using inoculum quantities o f 1 g (ml), 0.1 g (ml), and 0.01 g (ml) recommended by the FSIS, were obtained.
Statistical analysis. Salmonella prevalence and concentration data were subjected to Pearson’s chi-square test of DPS software (DPS 9.5, Institute of Insect Science, Zhejiang University, Hangzhou, China) to determine the significant variation, if any, among different months, seasons (spring, summer, autumn, and winter), market types (supermarket and wet market), chicken types (Sanhuang chicken, white chicken, black-bone chicken, and native chicken), and storage temperature (frozen, chilled, and ambient). For the statistical analysis and study, the seasons were defined as follows: spring (March, April, and May); summer (June, July, and August); autumn (September, October, and November); and winter (December, January, and February). The results were considered significant at the 5% (with a = 0.05) level to evaluate whether there is any association between the prevalence and concentration of Salmonella. RESULTS
The overall prevalence of Salmonella for chicken samples (n = 240) was 43.3%; the MPN values ranged from 0.0036 to 0.8596 MPN per g, and the average MPN value was 0.1655 MPN per g, except in 11 samples (4.58%), which had an MPN value of more than 11 MPN per g (Table 1). The contamination rate and MPN value exhibited a similar tendency by time during the period of surveillance (Table 1). In January and February, the contamination rates of retail chicken were both 10%, and the MPN value was similar as well. From February to April, both the contamination rate and MPN value increased sharply from 10 to 80% and from 0.0062 to 0.8596 MPN per g, respectively. The period from May to September experienced a slight fall in the number of the Salmonellapositive samples, and MPN values compared with those of April. Flowever, they were still at a relatively high level: from 50 to 70% and from 0.2084 to 0.3948 MPN per g, respectively, except the MPN value of June. From September to the end of the year, both the prevalence rate
1960
J. Food Prot., Vol. 76, No. 11
WANG ET AL.
TABLE 2. Prevalence o f Salmonella on chilled, frozen, and fi’esh chicken carcasses from two types o f markets Markets type
Storage condition
No. of samples examined
No. (%) of Salmonellapositive samples"
Avg MPN value (MPN/g)
No. (%) of MPN value >11 (MPN/g)
Supennarket
Chilled Frozen Freshly slaughtered
60 60 120
27 (45) A 32 (53.3) 45 (37.5)
0.0988 0.1804 0.1912
3 (5 ) 4 (6.67) 4 (3.33)
240
104 (43.3)
0.1655
11 (4.58)
Wet market Total
a a
a Within a column, data labeled with letters indicate not significantly different (P > 0.05).
and MPN value declined sharply. The peak of Salmonella prevalence rate and MPN value were both seen in April 2011; the lowest prevalence rate and MPN value were found in November 2011 (Table 1). The MPN values of more than 11 MPN per g were observed in 11 samples (4.58%), indicating that these chickens were seriously contaminated by Salmonella. Among these 11 samples, 4 were collected in May, 3 were collected in July, 2 in April, and 1 in September of 2011 and 1 in March of 2012. The high prevalence and MPN value of Salmonella in chicken were detected in late spring, summer, and early autumn (from March to September; Table 1). The contamination rate and MPN value of Salmonella both declined from spring to winter. The contamination rate in spring (73.3%) and summer (58.3%) were significantly higher (P < 0.05) than that in autumn (27.7%) and in winter (13.3%; Table 1). The tendency of the MPN value during the whole year was similar to the prevalence rate. The highest MPN value (0.3948 MPN per g) was detected in spring, followed by that in summer (0.2123 MPN per g) and autumn (0.0733 MPN per g), with the lowest value in winter (0.0093 MPN per g). Different seasons significantly (P < 0.05) affected the concentrations of Salmonella in chicken (Table 1). The average MPN values of Salmonella recovered from chilled and frozen chicken collected in the supermarket were 0.0988 and 0.1804 MPN per g, respectively, and the average MPN value of Salmonella from freshly slaughtered chicken in the wet market was 0.1912 MPN per g. No significant differences (P > 0.05) were found among them. Mean while, no significant difference (P > 0.05) was observed among contamination rates of chilled chicken (45.0%) and frozen chicken (53.3%) collected in the supennarket and that of freshly slaughtered chicken in wet markets (37.5%; Table 2). Regarding the type of market, the highest prevalence rate (53.5%) of Salmonella was found in the frozen chicken collected in the supermarket, while the
highest MPN value (0.1912 MPN per g) of Salmonella was found within the freshly slaughtered chicken collected from the wet market. The prevalence and MPN value of Salmonella in chicken carcasses of Sanhuang chicken, white chicken, black-bone chicken, and native chicken varied within the type of chickens (Table 3). The overall Salmonella prevalence in Sanhuang chicken, white chicken, black-bone chicken, and native chicken was 58.1% (n = 31), 49.5% (n = 95), 38.9% (n = 54), and 30.3% (n = 60), respectively. Chicken type did not significantly (P > 0.05) affect Salmonella prevalence. The overall MPN value of Salmonella in Sanhuang chicken, white chicken, black-bone chicken, and native chicken were 0.2329, 0.2511, 0.0930, and 0.6496 MPN per g, respectively. No significant differences (P > 0.05) were observed within the type of chicken, with MPN values of more than 11 MPN per g (Table 3). Among the 11 samples that had MPN values beyond 11 MPN per g, 6 were white chicken and 3 were black-bone chicken (Table 3). DISCUSSION Chicken and chicken products are widely known to be an important reservoir for Salmonella, and they have been ascribed as vehicles of Salmonella infections (1). The incidence of Salmonella in chicken carcasses in most countries ranged from 20 to 70% (3), which was in agreement with our study (43.3%). In an earlier study conducted in the United States, approximately 22% (n = 378) of raw chicken meats were contaminated with Salmonella (8). Other surveys revealed that the prevalence of Salmonella was 68.2% (n = 301) on both raw chicken meat and chicken giblets in Addis Ababa, Ethiopia (12), 66% (n = 705) on chicken meat in Bangkok, Thailand (6), 38% (n = 100) on chicken meat in Hyderabad, Pakistan (10), 34% (n = 200) on packaged fresh raw chicken in Anatolia (17), 45% (n = 190) on chicken in Tehran, Iran
TABLE 3. Prevalence o f Salmonella in four types o f chicken samples Chicken type
No. of samples examined
Sanhuang chicken White chicken Black-bone chicken Native chicken
31 95 54 60
Total
240
No. (%) of Salmonella positive samples" 18 47 21 18
(58.1) (49.5) (38.9) (30.0)
Avg MPN value (MPN/g)
No. (%) of MPN value >11 (MPN/g)
0.2329 0.2510 0.0913 0.0661
1 (3.23) 6 (6.32) 3 (5.56) 1 (1.67)
0.1655
11 (4.58)
a a a a
104 (43.3)
" Within a column, data labeled with letters indicate not significantly different (P > 0.05).
J. Food Prot., Vo!. 76, No. 11
PREVALENCE AND MPN OF SALMONELLA IN RETAIL CHICKEN
(3). The level of Salmonella contamination of chicken samples observed in our study (43.3%) was similar to the findings of the previous studies. The variation in Salmonella prevalence could be partly due to differences in the number of samples collected, sample type, chicken breed, sampling procedures and the detection methods used (9). Based on the results observed in the present investiga tion, it was obvious that the season was one of the important factors influencing the prevalence of Salmonella. For example, concentrations and prevalence of Salmonella in spring and summer were significantly higher than that in autumn and winter. Our results were in accordance with previous investigations showing a higher probability of chickens being infected during the summer (18). Wet and hot seasons enhance the survival of Salmonella, providing favorable growth conditions for Salmonella, and conse quently increasing its persistence. Salmonella concentration is a particular consumer safety concern, Salmonella, even at low concentrations, can multiply to hazardous levels if the right conditions emerge, for example, by temperature abuse (5). So far, however, few quantitative concentration data on Salmonella have been reported for food animals. A Belgian study found that 1.3% of Salmonella-positive cutting meat samples were contaminated with at least 10 CFU/g Salmonella and 7.4% with 1 to 10 CFU/g Salmonella (4). In a more recent Cambodian study, a very high level of contamination of Salmonella was found at 3 to 4 log CFU/g for 22.4% of samples (7). A positive correlation between the contamination rates of Salmonella and the corresponding average MPN values was found in our study. Salmonella was found in retail chicken collected from both supermarket and wet markets. The present study indicated that chilled and frozen chicken from supermarkets carried more Salmonella than freshly slaughtered chicken from wet markets, which agreed with the findings by Capita et al. (2), who found that the rate of Salmonellacontaminated carcasses from supermarkets (75%) was higher than that from wet markets (25%), and is possibly due to the short time that chicken carcasses were kept in wet markets (generally less than 16 h) (2). In contrast, Yang et al. (16) observed that Salmonella prevalence was the highest in wet markets (54.4%) compared with large markets (50.3%) and small markets (52.1%), with the possible causes of the difference being that the supply of potable water was limited in the wet market and eviscerated birds were usually rinsed with minimal amounts of water or dipped in a tank without frequent changes of water. On the contrary, the highest average MPN value of Salmonella was found in the freshly slaughtered chickens collected in wet markets, followed by the frozen samples and chilled ones sampled from supermarkets, possibly because of the employment of “ use by” dates in the supermarkets and also the packaging, which prevents further cross-contami nation between chicken carcasses. Moreover, most of the wet markets do not operate in a safe and clean environment: the sellers rarely place the appropriate covering over displayed carcasses, and they often use the same cutting knife for the uninfected and infected carcasses. Addition ally, the sellers at these wet markets were so busy that they
1961
seldom had time to wash their hands, scales, and other tools, which together could lead to a higher chance of cross contamination (16). The study indicated that chicken carcasses at the retail level could be considered a major potential source of human salmonellosis. Consequently, efforts should be made to inform consumers to follow the basic instructions regarding storage temperature, cooking, and prevention of contami nation and cross-contamination. Moreover, one urgent need is that food handlers be trained and educated in microbial risks associated with poultry meat and how to control them. The current study provided important data on Salmonella concentrations and incidence for the food safety risk assessment of retail chicken. Further characterization, including serotype, genotype, and antimicrobial susceptibil ity of these Salmonella isolates, should be conducted in the future. ACKNOWLEDGMENTS This study was financially supported by the Centers for Disease Control and Prevention, Ministry of Health of the People’s Republic of China. We also appreciate Dr. Shuangkui Du for his helpful discussions and advice on statistical analysis.
REFERENCES 1. 2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Baumler, A. J., B. M. Hargis, and R. M. Tsolis. 2000. Tracing the origins of Salmonella outbreaks. Science 287:50-52. Capita, R., M. Alvarez-Astorga, C. Alonso-Calleja, B. Moreno, and M. del Camino Garcia-Femandez. 2003. Occurrence of salmonellae in retail chicken carcasses and their products in Spain. Int. J. Food Microbiol. 81:169-173. Dallal, M. S. D., M. P. Doyle, M. Rezadehbashi, H. Dabiri, M. Sanaei, S. Modarresi, R. Bakhdari, K. Sharifiy, M. Taremi, M. R. Zali, and M. K. Sharifi-Yazdi. 2010. Prevalence and antimicrobial resistance profiles of Salmonella serotypes, Campylobacter and Yersinia spp. isolated from retail chicken and beef, Tehran, Iran. Food Control 21:388-392. Ghafir, Y., B. China, N. Korsak, K. Dierick, J. M. Collard, C. Godard, De Z. Lieven, and D. Georges. 2005. Belgian surveillance plans to assess changes in Salmonella prevalence in meat at different production stages. J. Food Prot. 68:2269-2277. Hansen, T. B„ B. B. Christensen, and S. Aabo. 2010. Salmonella in pork cuttings in supermarkets and butchers’ shops in Denmark in 2002 and 2006. Zoonoses Public Health 57:3-29. Jemgklinchani, J., C. Koowatananukuv, K. Daingprom, and K. Saitanu. 1994. Occurrence of salmonellae in raw broilers and their products in Thailand. J. Food Prot. 57:808-810. Lay, K. S., Y. Vuthy, P. Song, and J. L. Sarthou. 2011. Prevalence, numbers and antimicrobial susceptibilities of Salmonella serovars and Campylobacter spp. in retail poultry in Phnom Penh, Cambodia. J. Vet. Med. Sci. 73:325-329. M’ikanatha, N. M., C. H. Sandt, A. R. Localio, D. Tewari, S. C. Rankin, J. M. Whichard, S. F. Altekruse, E. Lautenbach, J. P. Folster, A. Russo, T. M. Chiller, S. M. Reynolds, and P. F. McDermott. 2010. Multidrug-resistant Salmonella isolates from retail chicken meat compared with human clinical isolates. Foodborne Pathog. Dis. 7: 929-934. Molla, B., and A. Mesfin. 2003. A survey of Salmonella contamination in chicken carcasss and giblets in Central Ethiopia. Rev. Med. Vet. 154:267-270. Soomro, A. H., M. Khaskheli, M. B. Bhutto, G. Shah, A. Memon, and P. Dewani. 2010. Prevalence and antimicrobial resistance of Salmonella serovars isolated from poultry meat in Hyderabad, Pakistan. Turk. J. Vet. Anim. Sci. 34:455^160. Stevens, A., Y. Kabore, J. D. Perrier-Gros-Claude, Y. Millemann. A. Brisabois, M. Catteau, J. F. Cavin, and B. Dufour. 2006. Prevalence
1962
12.
13.
14.
15.
J. Food Prot., Vol. 76, No. 11
WANG ET AL.
and antibiotic-resistance of Salmonella isolated from beef sampled from the slaughterhouse and from retailers in Dakar (Senegal). Ini. J. Food Microbiol. 110:178-186. Tibaijuka, B„ B. Molla, G. Hildebrandt, and J. Kleer. 2003. Occurrence of Salmonellae in retail raw chicken products in Ethiopia. Berl. Muench. Tieraerztl. Wochenschr. 116:55-58. U.S. Department of Agriculture, Food Safety and Inspection Service. 2008 laboratory guidebook: most probable number procedure and tables. Available at: http://www.fsis.usda.gov/PDF/MLG_Appendix_ 2_03.pdf. Accessed 24 January 2011. Wegener, H. C., T. Hald, L. F. Wong, M. Mogens, H. Korsgaard, F. Bager, P. Gemer-Smidt, and K. Molbak. 2003. Salmonella control programs in Denmark. Emerg. Infect. Dis. 9:774-780. World Health Organization. 2007. Salmonella control at the source. International Food Safety Authorities Network (INFOSAN). p. 1-4.
16.
17.
18.
19.
Available at: http://www.who.int/foodsafety/fs_management/No_03_ salmonella_May07_en.pdf. Accessed 10 January 2013. Yang, B„ M. Xi, X. Wang, S. Cui, T. Yue, H. Hao, Y. Wang, Y. Cui, W. Q. Alali, J. Meng, W. Isabel, D. M. Lo Fo Wong, and M. D. Doyle. 2011. Prevalence of Salmonella on raw poultry at retail markets in China. J. Food Prot. 74:1724—1728. Yildirim, Y., Z. Gonulalan, S. Pamuk, and N. Ertas. 2011. Incidence and antibiotic resistance of Salmonella spp. on raw chicken carcasses. Food Res. Int. 44:725-728. Zdragas, A., K. Mazaraki, G. Vafeas, V. Giantzi, T. Papadopoulos, and L. Ekateriniadou. 2012. Prevalence, seasonal occurrence and antimicrobial resistance of Salmonella in poultry retail products in Greece. Lett. Appl. Microbiol. 55:308-313. Zweifel, C., and R. Stephan. 2012. Spices and herbs as source of Salmonella-related foodbome diseases. Food Res. Int. 45:765-769.
Copyright of Journal of Food Protection is the property of Allen Press Publishing Services Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Research Note
Prevalence and Quantitative Detection of Salm onella in Retail Raw Chicken in Shaanxi, China JIAQI WANG,1 HAIYUN WU, 12 MIAO SONG,1 FENGQIN LI,3 JIANGHUI ZHU,3 MEILI XI,1 XIN WANG, 1 XIAODONG XIA, 1 JIANGHONG MENG, 14 BAOWEI YANG,1* AND SHENGHUI C M 5 1College o f Food Science and Engineering, Northwest A&F University, Shaanxi, 712100, China; 2School o f Pharmacy, Shanghai Jiao Tong University, Shanghai, 200240, China; 3China National Center fo r Food Safety Risk Assessment, Beijing, 100021, China; 4Joint Institute for Food Safety and Applied Nutrition, and Department o f Nutrition and Food Science, University o f Maryland, College Park, Maryland 20742, USA; and ^National Institutes fo r Food and Drug Control, Beijing, 100050, China MS 13-168: Received 26 April 2013/Accepted 28 June 2013
A BSTR A CT Quantitative Salmonella concentrations and prevalence of Salmonella in raw chicken at the retail level in Shaanxi province, China, was determined in this study. Two hundred forty samples were collected in Yangling and the surrounding cities, in Shaanxi Province, China, for data collection over 12 successive months from April 2011 to March 2012. During the whole surveillance year, the overall Salmonella contamination rate of retail raw chicken was identified as 43.3%, the mostprobable-number (MPN) values ranged from 0.0036 to 0.8596 MPN per g, and the average value was 0.1655 MPN per g, except in 11 of the samples, which had MPN values of more than 11 MPN per g. In April 2011, the highest prevalence rate, which was 80.0%, was observed, and the corresponding MPN value was 0.8596 MPN per g. Observed in November 2011, the lowest prevalence rate and the corresponding MPN value were 5.0% and 0.0036 MPN per g, respectively. Prevalence of Salmonella in chicken collected from the supermarket (49.2%) was higher but not significantly different from that in the wet markets (37.5%), although the MPN value of samples in the wet market (0.1912 MPN per g) was higher than that in the supermarket (0.1396 MPN per g). Prevalence of Salmonella was the highest in the frozen chicken (53.3%) compared with chilled chicken (45.0%) and freshly slaughtered chicken (37.5%); however, the MPN value of the freshly slaughtered chicken (0.1912 MPN per g) was higher than those of either frozen chicken (0.1804 MPN per g) or the chilled chicken (0.0988 MPN per g).
Salmonella is a genus of gram-negative rod-shaped bacteria of the Enterobacteriaceae family and can cause a wide range of human diseases such as gastroenteritis and enteric fever. The natural habitat of Salmonella is in the intestinal tract of animals and humans (19). Poultry and poultry products are usually incriminated in outbreaks of human salmonellosis (9). Chicken is widely acknowledged as a significant reservoir for Salmonella and has frequently been regarded as a source of Salmonella contamination (2). Salmonella often reaches the carcasses from the intestinal tracts or fecal materials of poultry (9). A link between the prevalence of Salmonella infection in poultry and human cases of salmonellosis was identified in a risk assessment conducted by the Food and Agriculture Organization of the United Nations and the World Health Organization (15). This report indicated that reducing rates of Salmonella infection in poultry flocks would lead to a direct reduction of salmonellosis within that community (15). Thus, surveillance of the level of Salmonella contam ination in food and food processing environments is * Author for correspondence. Tel: +86-29-87092486; Fax: 87092486; E-mail: [email protected].
+ 86-29-
necessary to control the spread of this pathogen from food to human. Control of Salmonella in food animals has been successful in the United States (8), Sweden, and Denmark (14) and has led to low levels of salmonellosis in these countries. The true incidence of salmonellosis in both humans and animals is difficult to evaluate in developing countries because of the lack of epidemiological surveil lance systems (12). In China, most research on Salmonella focused on the prevalence in different food animals, although few studies determined the quantitative Salmonella concentrations in food samples. The aim of the study was to determine the prevalence and quantitative concentrations of Salmonella in raw poultry at the retail level in Shaanxi province, China, and to collect data for quantitative risk assessment of Salmonella that was carried out by the China National Center for Food Safety Risk Assessment. M A TER IA LS A N D M E TH O D S Sample collection. For a total of 240 samples, 20 chicken carcasses, including 10 freshly slaughtered chickens in a wet market and 5 chilled and 5 frozen chicken samples in a supermarket, respectively, were collected monthly in Yangling and the surrounding cities in Shaanxi Province, China. After collection, each sample was placed aseptically in a sterile plastic
J. Food Prat., Vol. 76, No. 11
PREVALENCE AND MPN OF SAIMONF.UA IN RETAIL CHICKEN
1959
TABLE 1. Prevalence o f Salmonella in retail chicken by month Sampling date
No. of samples examined
April 2011 May 2011 June 2011 July 2011 August 2011 September 2011 October 2011 November 2011 December 2011 January 2012 February 2012 March 2012
20 20 20 20 20 20 20 20 20 20 20 20
Total
240
No. (%) of Salmonellapositive samples0 16 14 10 13 12 13 3 1 4 2 2 14
(80.0) B (70.0) b e (50.0) ABDE (65.0) BCE (60.0) ABCE (65.0) b c e (15.0) ACD (5.0) d (20.0) ADE (10.0) AD (10.0) AD (70.0) BE
104 (43.3)
Avg MPN value (MPN/g)
No. (%) of MPN value >11 (MPN/g)
0.8596 0.2453 0.0329 0.2084 0.3948 0.2104 0.0128 0.0036 0.0171 0.0046 0.0062 0.0805
2 (1 0 ) 4 (20) 0 (0 ) 3 (15) 0 (0) 1 (5) 0 (0 ) 0 (0 ) 0 (0 ) 0 (0 ) 0 (0 ) 1 (5)
0.1655
11 (4.58)
a Within a column, data labeled with different letters indicate significantly different (P < 0.05). bag. After it was marked, the sample was immediately stored in an icebox and transported to the Microbial Food Safety Laboratory at Northwest A&F University (Yangling, Shaanxi, China). The time from collection to isolation was no more than 4 h to avoid the potential growth of Salmonella during sample storage and transportation. The chicken type was acquired from the package labels or from the retailer.
Microbiological analysis. Analytical testing for Salmonella was carried out in the Microbial Food Safety Laboratory in Northwest A&F University. The analytical procedure for Salmo nella detection and most-probable-number (MPN) enumeration followed the method suggested by Food Safety and Inspection Service (FSIS) of the U.S. Department of Agriculture and China National Center for Food Safety Risk Assessment method (13). Briefly, each sample was weighed after transportation to the laboratory, then 500-fold of sterilized buffered peptone water (BPW; Difco, BD, Sparks, MD), according to the weight (kilograms), was added to the chicken carcass. The chicken was rinsed by manually massaging and using a rocking motion for 5 min to ensure that the surface, internal, and external layers of the chicken carcass would contact the rinse, and the rinse solution was used for the Salmonella concentration test. Three empty tubes were set up, as well as three tubes containing 9 ml of BPW and three tubes containing 9.9 ml of BPW. Ten milliliters of the chicken rinse solution was added to three empty tubes (representing 1 g), and then 1 ml of the rinse was transferred to each of three tubes that contained 9 ml of BPW (representing 0.1 g) and 100 pi of the rinse to each of three tubes that contained 9.9 ml of BPW (representing 0.01 g), respectively. If needed, 10-fold serial dilutions representing 0 .0 0 1-g samples were made. After inoculation, the tubes were preenriched for 20 to 24 h at 35 + 2°C. A portion (0.5 + 0.05 ml) of the preenrichment culture was transferred to 10 ml of tetrathionate broth (Difco) and 0.1 ± 0.02 ml into 10 ml of modified Rappaport-Vassiliadis broth (Difco), then incubated at 42 + 0.5°C with shaking at 100 rpm for 22 to 24 h. After incubation, a loop (diameter: 3 mm) of tetrathionate broth and modified Rappaport-Vassiliadis broth culture was streaked onto xylose lysine Tergitol 4 agar (Difco) and incubated at 35 + 2°C for 22 to 24 h. Colonies of presumptive Salmonella on xylose lysine Tergitol 4 plates were selected and purified on MacConkey agar; isolates with typical Salmonella phenotypes were confirmed by using the API 20E test kit (bioMerieux, Inc., Hazelwood, MO).
The MPN of each sample was acquired from the table of the MPN index, and 95% confidence limits for various combinations of positive tubes in a three-tube dilution series, using inoculum quantities o f 1 g (ml), 0.1 g (ml), and 0.01 g (ml) recommended by the FSIS, were obtained.
Statistical analysis. Salmonella prevalence and concentration data were subjected to Pearson’s chi-square test of DPS software (DPS 9.5, Institute of Insect Science, Zhejiang University, Hangzhou, China) to determine the significant variation, if any, among different months, seasons (spring, summer, autumn, and winter), market types (supermarket and wet market), chicken types (Sanhuang chicken, white chicken, black-bone chicken, and native chicken), and storage temperature (frozen, chilled, and ambient). For the statistical analysis and study, the seasons were defined as follows: spring (March, April, and May); summer (June, July, and August); autumn (September, October, and November); and winter (December, January, and February). The results were considered significant at the 5% (with a = 0.05) level to evaluate whether there is any association between the prevalence and concentration of Salmonella. RESULTS
The overall prevalence of Salmonella for chicken samples (n = 240) was 43.3%; the MPN values ranged from 0.0036 to 0.8596 MPN per g, and the average MPN value was 0.1655 MPN per g, except in 11 samples (4.58%), which had an MPN value of more than 11 MPN per g (Table 1). The contamination rate and MPN value exhibited a similar tendency by time during the period of surveillance (Table 1). In January and February, the contamination rates of retail chicken were both 10%, and the MPN value was similar as well. From February to April, both the contamination rate and MPN value increased sharply from 10 to 80% and from 0.0062 to 0.8596 MPN per g, respectively. The period from May to September experienced a slight fall in the number of the Salmonellapositive samples, and MPN values compared with those of April. Flowever, they were still at a relatively high level: from 50 to 70% and from 0.2084 to 0.3948 MPN per g, respectively, except the MPN value of June. From September to the end of the year, both the prevalence rate
1960
J. Food Prot., Vol. 76, No. 11
WANG ET AL.
TABLE 2. Prevalence o f Salmonella on chilled, frozen, and fi’esh chicken carcasses from two types o f markets Markets type
Storage condition
No. of samples examined
No. (%) of Salmonellapositive samples"
Avg MPN value (MPN/g)
No. (%) of MPN value >11 (MPN/g)
Supennarket
Chilled Frozen Freshly slaughtered
60 60 120
27 (45) A 32 (53.3) 45 (37.5)
0.0988 0.1804 0.1912
3 (5 ) 4 (6.67) 4 (3.33)
240
104 (43.3)
0.1655
11 (4.58)
Wet market Total
a a
a Within a column, data labeled with letters indicate not significantly different (P > 0.05).
and MPN value declined sharply. The peak of Salmonella prevalence rate and MPN value were both seen in April 2011; the lowest prevalence rate and MPN value were found in November 2011 (Table 1). The MPN values of more than 11 MPN per g were observed in 11 samples (4.58%), indicating that these chickens were seriously contaminated by Salmonella. Among these 11 samples, 4 were collected in May, 3 were collected in July, 2 in April, and 1 in September of 2011 and 1 in March of 2012. The high prevalence and MPN value of Salmonella in chicken were detected in late spring, summer, and early autumn (from March to September; Table 1). The contamination rate and MPN value of Salmonella both declined from spring to winter. The contamination rate in spring (73.3%) and summer (58.3%) were significantly higher (P < 0.05) than that in autumn (27.7%) and in winter (13.3%; Table 1). The tendency of the MPN value during the whole year was similar to the prevalence rate. The highest MPN value (0.3948 MPN per g) was detected in spring, followed by that in summer (0.2123 MPN per g) and autumn (0.0733 MPN per g), with the lowest value in winter (0.0093 MPN per g). Different seasons significantly (P < 0.05) affected the concentrations of Salmonella in chicken (Table 1). The average MPN values of Salmonella recovered from chilled and frozen chicken collected in the supermarket were 0.0988 and 0.1804 MPN per g, respectively, and the average MPN value of Salmonella from freshly slaughtered chicken in the wet market was 0.1912 MPN per g. No significant differences (P > 0.05) were found among them. Mean while, no significant difference (P > 0.05) was observed among contamination rates of chilled chicken (45.0%) and frozen chicken (53.3%) collected in the supennarket and that of freshly slaughtered chicken in wet markets (37.5%; Table 2). Regarding the type of market, the highest prevalence rate (53.5%) of Salmonella was found in the frozen chicken collected in the supermarket, while the
highest MPN value (0.1912 MPN per g) of Salmonella was found within the freshly slaughtered chicken collected from the wet market. The prevalence and MPN value of Salmonella in chicken carcasses of Sanhuang chicken, white chicken, black-bone chicken, and native chicken varied within the type of chickens (Table 3). The overall Salmonella prevalence in Sanhuang chicken, white chicken, black-bone chicken, and native chicken was 58.1% (n = 31), 49.5% (n = 95), 38.9% (n = 54), and 30.3% (n = 60), respectively. Chicken type did not significantly (P > 0.05) affect Salmonella prevalence. The overall MPN value of Salmonella in Sanhuang chicken, white chicken, black-bone chicken, and native chicken were 0.2329, 0.2511, 0.0930, and 0.6496 MPN per g, respectively. No significant differences (P > 0.05) were observed within the type of chicken, with MPN values of more than 11 MPN per g (Table 3). Among the 11 samples that had MPN values beyond 11 MPN per g, 6 were white chicken and 3 were black-bone chicken (Table 3). DISCUSSION Chicken and chicken products are widely known to be an important reservoir for Salmonella, and they have been ascribed as vehicles of Salmonella infections (1). The incidence of Salmonella in chicken carcasses in most countries ranged from 20 to 70% (3), which was in agreement with our study (43.3%). In an earlier study conducted in the United States, approximately 22% (n = 378) of raw chicken meats were contaminated with Salmonella (8). Other surveys revealed that the prevalence of Salmonella was 68.2% (n = 301) on both raw chicken meat and chicken giblets in Addis Ababa, Ethiopia (12), 66% (n = 705) on chicken meat in Bangkok, Thailand (6), 38% (n = 100) on chicken meat in Hyderabad, Pakistan (10), 34% (n = 200) on packaged fresh raw chicken in Anatolia (17), 45% (n = 190) on chicken in Tehran, Iran
TABLE 3. Prevalence o f Salmonella in four types o f chicken samples Chicken type
No. of samples examined
Sanhuang chicken White chicken Black-bone chicken Native chicken
31 95 54 60
Total
240
No. (%) of Salmonella positive samples" 18 47 21 18
(58.1) (49.5) (38.9) (30.0)
Avg MPN value (MPN/g)
No. (%) of MPN value >11 (MPN/g)
0.2329 0.2510 0.0913 0.0661
1 (3.23) 6 (6.32) 3 (5.56) 1 (1.67)
0.1655
11 (4.58)
a a a a
104 (43.3)
" Within a column, data labeled with letters indicate not significantly different (P > 0.05).
J. Food Prot., Vo!. 76, No. 11
PREVALENCE AND MPN OF SALMONELLA IN RETAIL CHICKEN
(3). The level of Salmonella contamination of chicken samples observed in our study (43.3%) was similar to the findings of the previous studies. The variation in Salmonella prevalence could be partly due to differences in the number of samples collected, sample type, chicken breed, sampling procedures and the detection methods used (9). Based on the results observed in the present investiga tion, it was obvious that the season was one of the important factors influencing the prevalence of Salmonella. For example, concentrations and prevalence of Salmonella in spring and summer were significantly higher than that in autumn and winter. Our results were in accordance with previous investigations showing a higher probability of chickens being infected during the summer (18). Wet and hot seasons enhance the survival of Salmonella, providing favorable growth conditions for Salmonella, and conse quently increasing its persistence. Salmonella concentration is a particular consumer safety concern, Salmonella, even at low concentrations, can multiply to hazardous levels if the right conditions emerge, for example, by temperature abuse (5). So far, however, few quantitative concentration data on Salmonella have been reported for food animals. A Belgian study found that 1.3% of Salmonella-positive cutting meat samples were contaminated with at least 10 CFU/g Salmonella and 7.4% with 1 to 10 CFU/g Salmonella (4). In a more recent Cambodian study, a very high level of contamination of Salmonella was found at 3 to 4 log CFU/g for 22.4% of samples (7). A positive correlation between the contamination rates of Salmonella and the corresponding average MPN values was found in our study. Salmonella was found in retail chicken collected from both supermarket and wet markets. The present study indicated that chilled and frozen chicken from supermarkets carried more Salmonella than freshly slaughtered chicken from wet markets, which agreed with the findings by Capita et al. (2), who found that the rate of Salmonellacontaminated carcasses from supermarkets (75%) was higher than that from wet markets (25%), and is possibly due to the short time that chicken carcasses were kept in wet markets (generally less than 16 h) (2). In contrast, Yang et al. (16) observed that Salmonella prevalence was the highest in wet markets (54.4%) compared with large markets (50.3%) and small markets (52.1%), with the possible causes of the difference being that the supply of potable water was limited in the wet market and eviscerated birds were usually rinsed with minimal amounts of water or dipped in a tank without frequent changes of water. On the contrary, the highest average MPN value of Salmonella was found in the freshly slaughtered chickens collected in wet markets, followed by the frozen samples and chilled ones sampled from supermarkets, possibly because of the employment of “ use by” dates in the supermarkets and also the packaging, which prevents further cross-contami nation between chicken carcasses. Moreover, most of the wet markets do not operate in a safe and clean environment: the sellers rarely place the appropriate covering over displayed carcasses, and they often use the same cutting knife for the uninfected and infected carcasses. Addition ally, the sellers at these wet markets were so busy that they
1961
seldom had time to wash their hands, scales, and other tools, which together could lead to a higher chance of cross contamination (16). The study indicated that chicken carcasses at the retail level could be considered a major potential source of human salmonellosis. Consequently, efforts should be made to inform consumers to follow the basic instructions regarding storage temperature, cooking, and prevention of contami nation and cross-contamination. Moreover, one urgent need is that food handlers be trained and educated in microbial risks associated with poultry meat and how to control them. The current study provided important data on Salmonella concentrations and incidence for the food safety risk assessment of retail chicken. Further characterization, including serotype, genotype, and antimicrobial susceptibil ity of these Salmonella isolates, should be conducted in the future. ACKNOWLEDGMENTS This study was financially supported by the Centers for Disease Control and Prevention, Ministry of Health of the People’s Republic of China. We also appreciate Dr. Shuangkui Du for his helpful discussions and advice on statistical analysis.
REFERENCES 1. 2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Baumler, A. J., B. M. Hargis, and R. M. Tsolis. 2000. Tracing the origins of Salmonella outbreaks. Science 287:50-52. Capita, R., M. Alvarez-Astorga, C. Alonso-Calleja, B. Moreno, and M. del Camino Garcia-Femandez. 2003. Occurrence of salmonellae in retail chicken carcasses and their products in Spain. Int. J. Food Microbiol. 81:169-173. Dallal, M. S. D., M. P. Doyle, M. Rezadehbashi, H. Dabiri, M. Sanaei, S. Modarresi, R. Bakhdari, K. Sharifiy, M. Taremi, M. R. Zali, and M. K. Sharifi-Yazdi. 2010. Prevalence and antimicrobial resistance profiles of Salmonella serotypes, Campylobacter and Yersinia spp. isolated from retail chicken and beef, Tehran, Iran. Food Control 21:388-392. Ghafir, Y., B. China, N. Korsak, K. Dierick, J. M. Collard, C. Godard, De Z. Lieven, and D. Georges. 2005. Belgian surveillance plans to assess changes in Salmonella prevalence in meat at different production stages. J. Food Prot. 68:2269-2277. Hansen, T. B„ B. B. Christensen, and S. Aabo. 2010. Salmonella in pork cuttings in supermarkets and butchers’ shops in Denmark in 2002 and 2006. Zoonoses Public Health 57:3-29. Jemgklinchani, J., C. Koowatananukuv, K. Daingprom, and K. Saitanu. 1994. Occurrence of salmonellae in raw broilers and their products in Thailand. J. Food Prot. 57:808-810. Lay, K. S., Y. Vuthy, P. Song, and J. L. Sarthou. 2011. Prevalence, numbers and antimicrobial susceptibilities of Salmonella serovars and Campylobacter spp. in retail poultry in Phnom Penh, Cambodia. J. Vet. Med. Sci. 73:325-329. M’ikanatha, N. M., C. H. Sandt, A. R. Localio, D. Tewari, S. C. Rankin, J. M. Whichard, S. F. Altekruse, E. Lautenbach, J. P. Folster, A. Russo, T. M. Chiller, S. M. Reynolds, and P. F. McDermott. 2010. Multidrug-resistant Salmonella isolates from retail chicken meat compared with human clinical isolates. Foodborne Pathog. Dis. 7: 929-934. Molla, B., and A. Mesfin. 2003. A survey of Salmonella contamination in chicken carcasss and giblets in Central Ethiopia. Rev. Med. Vet. 154:267-270. Soomro, A. H., M. Khaskheli, M. B. Bhutto, G. Shah, A. Memon, and P. Dewani. 2010. Prevalence and antimicrobial resistance of Salmonella serovars isolated from poultry meat in Hyderabad, Pakistan. Turk. J. Vet. Anim. Sci. 34:455^160. Stevens, A., Y. Kabore, J. D. Perrier-Gros-Claude, Y. Millemann. A. Brisabois, M. Catteau, J. F. Cavin, and B. Dufour. 2006. Prevalence
1962
12.
13.
14.
15.
J. Food Prot., Vol. 76, No. 11
WANG ET AL.
and antibiotic-resistance of Salmonella isolated from beef sampled from the slaughterhouse and from retailers in Dakar (Senegal). Ini. J. Food Microbiol. 110:178-186. Tibaijuka, B„ B. Molla, G. Hildebrandt, and J. Kleer. 2003. Occurrence of Salmonellae in retail raw chicken products in Ethiopia. Berl. Muench. Tieraerztl. Wochenschr. 116:55-58. U.S. Department of Agriculture, Food Safety and Inspection Service. 2008 laboratory guidebook: most probable number procedure and tables. Available at: http://www.fsis.usda.gov/PDF/MLG_Appendix_ 2_03.pdf. Accessed 24 January 2011. Wegener, H. C., T. Hald, L. F. Wong, M. Mogens, H. Korsgaard, F. Bager, P. Gemer-Smidt, and K. Molbak. 2003. Salmonella control programs in Denmark. Emerg. Infect. Dis. 9:774-780. World Health Organization. 2007. Salmonella control at the source. International Food Safety Authorities Network (INFOSAN). p. 1-4.
16.
17.
18.
19.
Available at: http://www.who.int/foodsafety/fs_management/No_03_ salmonella_May07_en.pdf. Accessed 10 January 2013. Yang, B„ M. Xi, X. Wang, S. Cui, T. Yue, H. Hao, Y. Wang, Y. Cui, W. Q. Alali, J. Meng, W. Isabel, D. M. Lo Fo Wong, and M. D. Doyle. 2011. Prevalence of Salmonella on raw poultry at retail markets in China. J. Food Prot. 74:1724—1728. Yildirim, Y., Z. Gonulalan, S. Pamuk, and N. Ertas. 2011. Incidence and antibiotic resistance of Salmonella spp. on raw chicken carcasses. Food Res. Int. 44:725-728. Zdragas, A., K. Mazaraki, G. Vafeas, V. Giantzi, T. Papadopoulos, and L. Ekateriniadou. 2012. Prevalence, seasonal occurrence and antimicrobial resistance of Salmonella in poultry retail products in Greece. Lett. Appl. Microbiol. 55:308-313. Zweifel, C., and R. Stephan. 2012. Spices and herbs as source of Salmonella-related foodbome diseases. Food Res. Int. 45:765-769.
Copyright of Journal of Food Protection is the property of Allen Press Publishing Services Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
